Design of diagnostic estimators for an automated guided vehicle

Abstract

The paper deals with the design of virtual sensors of longitudinal forces as well as torques acting onto an automated guided vehicle. These virtual sensors are based on a measurement set associated with accelerations, velocities and yaw rate sensors mounted on the robot. Subsequently, the estimated forces and torques are used for diagnostic purposes concerning their appropriate distributions. The main novelty of the proposed approach is that it eliminates the necessity of incorporating wheel models, which usually impairs the performance of the approaches proposed in the literature. The final part of the paper presents experimental results, which clearly exhibit the performance of the proposed approach.